Robust diagnostics in university performance studies

Abstract

The presence of anomalous observations (outliers) in a set of data is one of the greatest problems in methodological statistics, one that scientists were already aware of many years ago, as can be seen in the comments made by the American astronomer Peirce1 over 150 years ago.

In almost every true series of observations, some are found, which differ so much from the others as to indicate some abnormal source of error not contemplated in the theoretical discussions, and the introduction of which into the investigations can only serve to perplex and mislead the inquirer.

(Peirce, 1852)